基于仿射变换的快速全局运动估计算法

王思珺; 赵建; 韩希珍

doi:doi:10.3788/yjyxs20122702.0263

液晶与显示, 2012, 27 (2): 263, 网络出版: 2012-05-08

基于仿射变换的快速全局运动估计算法

Fast Global Motion Estimation Algorithm Based on Affine Transformation

论文大纲

王思珺 ^1,2,*赵建 ¹韩希珍 ¹

作者单位

¹ 中国科学院长春光学精密机械与物理研究所，吉林长春130033

² 中国科学院研究生院，北京100039

摘要

全局运动估计是计算机视觉、视频处理等领域广泛采用的手段之一。为了降低全局运动估计计算复杂度，提出一种基于像素梯度的全局运动估计方法。就运动模型而言，采用了复杂性与准确性较好均衡的六参数仿射模型。为了提高计算速度，采用了两层金字塔进行多分辨率计算，而且在每层迭代计算中都抽取梯度较大的特征点进行计算，同时结合了高斯-牛顿优化计算方法。实验验证了提出的全局运动估计算法的计算效率和精度。

Abstract

Global Motion Estimation is an important tool widely used in computer vision, video processing, and other fields. To reduce the computational complexity of global motion estimation, a global motion estimation method based on the pixel gradient strategy was proposed. With regard to global motion models, six-parameter affine model is adopted because of its reasonable tradeoff between complexity and accuracy. In order to accelerate the algorithm, two-level pyramid is utilized in the calculation. It uses those gradient bigger pixels in the calculation of every pyramid level, and combines Gauss-Newton algorithm. The extensive experiments show that the proposed method is efficient and robust in terms of both computational complexity and accuracy.

参考文献

[1] Mahmoud H, Bayoumi M. An efficient low-bit rate adaptive mesh-based motion compensation technique ［C］// Proc. 2nd Int. Workshop on Digital and Computational Video. Tampa, Florida, USA:IEEE Computer Society, 2001:164-172.

[2] Yosi K, Amir A. Fast gradient methods based on global motion estimation for video compression ［J］. IEEE Trans. Circuit Syst. Video Technol., 2003, 13(4):300-309.

[3] Dufaux F, Konrad J. Efficient, robust, and fast global motion estimation for video coding ［J］. IEEE Transactions on Image Processing, 2000, 9(3):497-501.

[4] 俞天力,章毓晋.基于全局运动信息的视频检索技术［J］.电子学报,2001,29(12A)：1794-1798.

[5] Rath G B, Makur A. Iterative least squares and compression based estimations for a four-parameter linear global motion model and global motion compensation ［J］. IEEE Transactions on Circuits and Systems for Video Techno-logy, 1999, 9(7):1075-1099.

[6] Bin Qi, Ghazal M, Amer A. Robust global motion estimation oriented to video object segmentation ［J］. IEEE Transactions on Image Processing, 2008,17(6):958-967.

[7] Keller Y, Averbuch A. Fast gradient methods based on global motion estimation for video compression ［J］. IEEE Transactions on Circuits and Systems for Video Technology, 2003, 13(4):300-309.

[8] Wang D, Wang L. Fast and robust algorithm for global motion estimation ［C］// Proc SPIE Conf on Visual Communications and Image Processing(VCIP), San Jose, CA, USA：SPIE 1997,3024:1144-1151.

[9] Li H, Liu G, Li Y, et al. A robust, efficient, and fast global motion estimation method from MPEG compressed video ［C］// IEEE Proc. PCM2002, LNCS 2532, Taiwan：SPIE， 2002: 151-158.

[10] Moscheni F, Dufaux F, Kunt M. A new two-stage global/local motion estimation based on a background/foreground segmentation ［C］//IEEE Proc. ICASSP95, Detroit, MI:IEEE,1995:2261-2264.

[11] Sorwar G, Murshed M, Dooley L. Fast global motion estimation using iterative least-square estimation technique ［C］// IEEE Proc. ICICSPCM 2003, Singapore, IEEE, 2003:282-286.

[12] 贺玉文,赵黎,钟玉琢,等.快速鲁棒的全局运动估计算法［J］.软件学报,2001,12(8):1220-1228.

[13] Press W H, Flannery B P, Teukolsky S A, et al. Numerical recipes in C: The Art of Scientific Computing ［M］. Cambridge: Cambridge University Press, 2003: 32-102.

王思珺, 赵建, 韩希珍. 基于仿射变换的快速全局运动估计算法[J]. 液晶与显示, 2012, 27(2): 263. WANG Si-jun, ZHAO Jian, HAN Xi-zhen. Fast Global Motion Estimation Algorithm Based on Affine Transformation[J]. Chinese Journal of Liquid Crystals and Displays, 2012, 27(2): 263.

基于仿射变换的快速全局运动估计算法

关于本站 Cookie 的使用提示

全站搜索

基于仿射变换的快速全局运动估计算法

相关论文

相关资讯

关于本站 Cookie 的使用提示

全站搜索